A Comparison of Various Correction and Blending Techniques for Creating an Improved Satellite-Gauge Rainfall Dataset over Australia

Satellites offer a way of estimating rainfall away from rain gauges which can be utilised to overcome the limitations imposed by gauge density on traditional rain gauge analyses. In this study, Australian station data along with the Japan Aerospace Exploration Agency’s (JAXA) Global Satellite Mapping of Precipitation (GSMaP) and the Bureau of Meteorology’s (BOM) Australian Gridded Climate Dataset (AGCD) rainfall analysis are combined to develop an improved satellite-gauge rainfall analysis over Australia that uses the strengths of the respective data sources. We investigated a variety of correction and blending methods with the aim of identifying the optimal blended dataset. The correction methods investigated were linear corrections to totals and anomalies, in addition to quantile-to-quantile matching. The blending methods tested used weights based on the error variance to MSWEP (Multi-Source Weighted Ensemble Product), distance to the closest gauge, and the error from a triple collocation analysis to ERA5 and Soil Moisture to Rain. A trade-off between away-from- and at-station performances was found, meaning there was a complementary nature between specific correction and blending methods. The most high-performance dataset was one corrected linearly to totals and subsequently blended to AGCD using an inverse error variance technique. This dataset demonstrated improved accuracy over its previous version, largely rectifying erroneous patches of excessive rainfall. Its modular use of individual datasets leads to potential applicability in other regions of the world.

Identification of Rainfall and Inundation pattern using remotely sensed data in Lake Sentarum Floodplain Area

Wetlands are vulnerable natural habitats that should be preserved to protect habitat for fish and wildlife, flood mitigation, improve water quality, recharge area, and maintain surface water flow during dry periods. Water bodies and swamp areas are two primary components of the wetland. Considering its essential roles for the ecosystem, Lake Sentarum was set as a national park area (Lake Sentarum National Park – TNDS), Indonesia's 15 national priority lakes, and; designated as a Ramsar site (The Convention on Wetlands) in Indonesia. Despite the significant roles for the ecosystem, providing the limnological characteristic of Lake Sentarum remains a challenge due to its remote location. This study aims to identify the rainfall and inundation characteristics in the Lake Sentarum area and develop the rainfall-inundation relationship in the TNDS area. First, we carried out rainfall analysis using the Climate Forecast System Reanalysis (CFSR) data. Second, we utilized a remote-sensing-based global surface water map from the Joint Research Centre (JRC) to describe the historical inundation pattern. Third, we applied the Normalized Different Water Index (NDWI) combined with Modification Normalized Different Water Index (MNDWI) to the selected Landsat dataset to extract the inundation area. Finally, we developed a rainfall-inundation relationship in the TNDS area. The result indicated that the yearly rainfall in the TNDS area has an increasing trend, with the highest peak in December and the second peak in April. Historical Landsat data shows that the TNDS has a complex pattern of inundation. The maximum water extent was 649 km2, with a 95 km2 as permanent (90>- 100 % water occurrence). These areas were constantly flooded, even in the dry season. The most significant non-permanent water was 161 km2 (80>- 90 % water occurrence). This permanent and larger temporary water area provides fish and other aquatic biotas habitats. It temporarily stores the water flowing slowly into the River Kapuas through the Tawang River. We captured the spatial inundation pattern and its relationship with the temporal regional rainfall. The developed relationship showed a lag of -60 days of accumulated rainfall correlated with the inundation area (R2 of 0.48, n=11). These findings will thus provide valuable data for lake managers and policy-makers to protect the biota and habitat in Lake Sentarum National Park area.

Sub-division wise probabilistic variability and extreme rainfall analysis of the Indian summer monsoon rainfall

Statistical analysis based on past data on probability of normal/excess rainfall, the probable future rainfall for smaller areas at different stages of the monsoon period will serve as an appropriate information system for efficient management of available surface water resource at the state and the national levels. In this paper the author have made an attempt in that direction and have brought out some important features of the monsoon rainfall. It is found that the probability of monthly rainfall becoming normal or excess is high in maximum number of sub-divisions in July and August and is least in September. It is further observed that the normality of rainfall as highly probable to the north of the monsoon trough in July and that to the south of the trough in August besides the west coast. The rainfall extreme values over a long period will be useful in determining the minimum assured and maximum probable future rainfall at different stages of the monsoon period. These information will be valuable to decision makers, managers etc. in their decision making process on real time basis.

Four statistical models for wet-spell analysis

Wet-spell analysis is an important part of rainfall analysis. The distribution of the length of wet-spells provides useful information on the temporal distribution of rainfall. This distribution has traditionally been modelled through different probability distributions. Here we compare four such models, namely, Cochran's model, truncated Poisson distribution, truncated negative binomial distribution, and logarithmic series distribution. These comparisons are accomplished with help of application to five rainguage stations in India.

EI-Nino southern oscillation and rainfall varIation over Bangladesh

In the present study, an attempt has been made to examine the variations of rainfall over Bangladesh and to find possible correlation with EI-Nino/Southern Oscillation (ENSO). Four stations have been chosen from four different climatic regions of Bangladesh for this purpose, namely Jessore, Dhaka. Barisal and Srimangal. The regions have been classified according to annual rainfall amounts. The rainfall data for forty three years, (1950-1992) have been analysed. The yearly mean rainfall shows a distinct negative decreasing tendency with the occurrence of ENSO.The seasonal rainfall analysis shows a somewhat better correlation.

Rainfall study for dry land areas of selected districts of Tamil Nadu for crop planning

Monthly rainfall data for 90 years from 1911 to 2000 and recent years weekly rainfall data from 1981 to 2010 for 30 years were collected from India Meteorological Department and used for analysis to propose alternate crops under dry land situation of three selected districts of Tamil Nadu (Dharmapuri, Tirunelveli and Villupuram). Based on the geographical location, the annual mean rainfall of the three selected districts did vary indicating spatial variability of rainfall in Tamil Nadu. The length of growing period was analysed employing three different methods namely Hargreave’s Moisture Availability Index, Jeevananda Reddy 14 weeks moving average and FAO water balance model. Conditional probability index was also computed for weeding and plant protection measures for the three districts. Under dry land situation, the shorter scale weekly rainfall analysis revealed 16 weeks growing period for Dharmapuri district and suggested crops were maize, ragi and mixture of sorghum, red gram and mochai. In respect of Tirunelveli district, the identified growing weeks were 12 and scientifically lesser Evapo-Transpiration requirement crops like pulses and minor millets had been suggested. For Villupuram district, the growing period was16 weeks and hence groundnut, sesame, maize, varagu and soybean had been suggested against cotton.

Long term rainfall trend over meteorological sub divisions and districts of India

In this paper the long term trend of annual and seasonal rainfall over different districts, Meteorological (Met.) sub-divisions and whole India have been studied using the long term rainfall data for the period from 1901 to 2013. The changes in amount and pattern of rainfall have a significant impact on agriculture, water resources management and overall economy of the country. Mann-Kendall test is applied to check the significance of the trend. Linear Regression and Theil-Sen’s non parametric test has been applied to estimate the trend. The study is carried out for 632 districts and 34 sub divisions of India by utilizing the gridded rainfall data (0.25° × 0.25°) over the main land except Andaman & Nicobar and Lakshadweep islands. Many authors have studied that extreme events are increasing but there is no trend in Pan India’s rainfall. It is observed from the annual rainfall analysis 10% of the number of districts are showing significant increasing trend and 13% significant decreasing (mainly in Uttar Pradesh) trend whereas irrespective of high and low rainfall regions, 10% area of the country is showing significant increasing trend and 8% of the area of the country showing significant decreasing trend in annual rainfall. In Meteorological Sub divisions, east & west UP are showing significant negative trend and some of the coastal sub divisions are showing positive trend. It is also observed that the country’s rainfall is not showing any trend.